In one conventional network arrangement, a communication stream is transmitted from a first host to a second host. The stream includes a plurality of packets. Software processes in the first host select the average timing of transmission or maximum latency of the packets so as to maintain a predictable data transmission rate for the packets. The second host includes a buffer to store the packets received from the first host.
In this conventional arrangement, the actual tinning of transmission of any given individual packet in the steam may vary, so long as the overall average timing of the transmission or maximum latency of the packets in the stream (i.e., taking into account all of the packets in the stream) conform to what has been selected for these parameters by the software processes in the first host. In order to prevent buffer overflow in the second host, the size of the buffer in the second host is selected so as to conform to the maximum buffering requirements that may be expected to prevail in a worse case communication scenario involving the stream (i.e., in a worst case communicate scenario given the average packet transmission timing or maximum packet latency selected by the software processes). This may result in inure buffer memory being allocated for packet storage than is desirable. This may make buffer memory usage and/or allocation less efficient.
Although the following Detailed Description will proceed with reference being made to illustrative embodiments, many alternatives, modifications, and variations thereof will be apparent to those skilled in the art. Accordingly, it is intended that the claimed subject matter be viewed broadly.